Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-3-27B-IT
    3. 40-GEMMASCOPE-2-RES-262K
    4. 16577
    Prev
    Next
    INDEX
    Explanations

    meta-information and self-awareness

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_40_width_262k_l0_medium
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    жду
    0.38
     अनिवार
    0.38
     rach
    0.37
     obligatorio
    0.37
    濃厚
    0.37
     mili
    0.37
    晹
    0.36
     ನ
    0.36
    輳
    0.36
     प्राकृतिक
    0.36
    POSITIVE LOGITS
     meta
    1.40
     Meta
    1.23
    Meta
    1.18
     metac
    1.16
    meta
    1.16
     мета
    1.04
     META
    1.03
     metaf
    1.00
    META
    0.96
     metas
    0.96
    Activations Density 0.022%

    No Known Activations